filters::KalmanFilterBase< stateDim, inputDim > Class Template Reference

Base class for Kalman Filters. More...

#include <KalmanFilterBase.h>

Inheritance diagram for filters::KalmanFilterBase< stateDim, inputDim >:

Collaboration diagram for filters::KalmanFilterBase< stateDim, inputDim >:

Public Member Functions
virtual void	predict (const Eigen::Matrix< double, inputDim, 1 > &input)=0
	Use the model to predict the next system state, given the current inputs.
void	reset ()
	Reset the filter.
void	reset (const Eigen::Matrix< double, stateDim, 1 > &state)
	Reset the filter.
void	reset (const Eigen::Matrix< double, stateDim, 1 > &state, const Eigen::Matrix< double, stateDim, 1 > &stdDevs)
	Reset the filter.
void	reset (const Eigen::Matrix< double, stateDim, 1 > &state, const Eigen::Matrix< double, stateDim, stateDim > &estCovMat)
	Reset the filter.
Eigen::Matrix< double, stateDim, 1 >	getState () const
	Gets the current state estimate.
Eigen::Matrix< double, stateDim, stateDim >	getEstimateCovarianceMat () const
	Get the current estimate covariance matrix.

Protected Attributes
Eigen::Matrix< double, stateDim, stateDim >	P
Eigen::Matrix< double, stateDim, 1 >	xHat

Detailed Description

template<int stateDim, int inputDim>
class filters::KalmanFilterBase< stateDim, inputDim >

Base class for Kalman Filters.

Template Parameters

stateDim	The dimension of the state vector.
inputDim	The dimension of the input vector, or system actions.

Member Function Documentation

getEstimateCovarianceMat()

template<int stateDim, int inputDim>

Eigen::Matrix< double, stateDim, stateDim > filters::KalmanFilterBase< stateDim, inputDim >::getEstimateCovarianceMat ( ) const

inline

Get the current estimate covariance matrix.

This is an indication of the uncertainty of the pose estimate.

Returns

The estimate covariance matrix, AKA the P matrix.

getState()

template<int stateDim, int inputDim>

Eigen::Matrix< double, stateDim, 1 > filters::KalmanFilterBase< stateDim, inputDim >::getState ( ) const

inline

Gets the current state estimate.

Returns

The state estimate.

predict()

template<int stateDim, int inputDim>

virtual void filters::KalmanFilterBase< stateDim, inputDim >::predict ( const Eigen::Matrix< double, inputDim, 1 > & input )

pure virtual

Use the model to predict the next system state, given the current inputs.

Parameters

input

The current inputs to the system.

Implemented in filters::ExtendedKalmanFilter< stateDim, inputDim, outputDim, processNoiseDim, outputNoiseDim >, and filters::KalmanFilter< stateDim, inputDim, outputDim >.

reset() [1/4]

template<int stateDim, int inputDim>

void filters::KalmanFilterBase< stateDim, inputDim >::reset ( )

inline

Reset the filter.

Sets the state estimate to the zero vector and resets the estimate covariance matrix to a diagonal matrix with large values.

reset() [2/4]

template<int stateDim, int inputDim>

void filters::KalmanFilterBase< stateDim, inputDim >::reset ( const Eigen::Matrix< double, stateDim, 1 > & state )

inline

Reset the filter.

Sets the state estimate to the supplied vector and resets the estimate covariance matrix to a diagonal matrix with large values.

Parameters

state

The state to which the state estimate will be set.

reset() [3/4]

template<int stateDim, int inputDim>

void filters::KalmanFilterBase< stateDim, inputDim >::reset ( const Eigen::Matrix< double, stateDim, 1 > & state, const Eigen::Matrix< double, stateDim, 1 > & stdDevs )

inline

Reset the filter.

Sets the state estimate to the supplied vector with the given uncertainty.

Parameters

state	The state to set the vector to.
stdDevs	The standard deviation of the measurement of each element in the state vector. This should not be the zero vector. If you want to represent high certainty, use very small values instead.

reset() [4/4]

template<int stateDim, int inputDim>

void filters::KalmanFilterBase< stateDim, inputDim >::reset ( const Eigen::Matrix< double, stateDim, 1 > & state, const Eigen::Matrix< double, stateDim, stateDim > & estCovMat )

inline

Reset the filter.

Sets the state estimate to the supplied vector and the estimate covariance matrix to the supplied matrix.

Parameters

state	The state to set the vector to.
estCovMat	The matrix to set the estimate covariance matrix to. This should not be the zero matrix. If you want to represent high certainty, use a diagonal matrix with very small values.

---

The documentation for this class was generated from the following file:

• src/filters/KalmanFilterBase.h